Product Catalog Hackathon with Ai Recap
In the beginning of August, data science teams at bol.com came together for a dynamic two-day datathon. Collaborating with different teams from different disciplines, their mission was to re-imagine how we discover similar products. This innovative approach involved combining the insights of text (titles/descriptions) and images. To achieve this, they converted both text and image data into vectors and then find similar products by for example using a Faiss index and other.
Product Catalog Hackathon with Ai Recap